No abstract available

[en] The reliability information which comes from subjective analysis is often incomplete prior. This information can be generally assumed to exist in the form of either a stated prior mean of R (reliability) or a stated prior credibility interval on R. An efficient approach is developed to determine a complete beta prior distribution from the subjective information according to the principle of maximum entropy, and the the reliability of survival/failure product is assessed via Bayes theorem. Numerical examples are presented to illustrate the methods

[en] In previous reports theoretical methods have been described for estimating the aerodynamic forces acting on cylinders vibrating laterally when surrounded by an annulus carrying high velocity gas. For a certain restricted set of geometries it is possible to predict whether a particular structure is stable or unstable and to determine the level of aerodynamic damping positive or negative due to the presence of the gas. This report describes experimental work which validates the computer program in which the theoretical methods are embodied; in particular the damping, inertial and decentralising forces acting on a cylinder in an annulus are measured and compared with theory over a range of frequencies from 0 to 25 Hz, and of Reynolds numbers from zero to 10⁴. In addition a summary of simple relationships is provided which can be used to provide credible initial estimates of both the positive and negative damping of cylinders in a range of annular geometries. (author)

[en] This technical note discusses the uses of personal opinion probabilities in addressing and resolving technical issues. The discussion raises questions without answering them. Are the assumptions underlying personal opinion probabilities use valid? For example, in the real world, are experts equal, unbiased, unemotional? Is there evidence to support the utility of personal opinion probabilities in contrast to other decision-making approaches in the absence of data (e.g. bounding analyses or conservative calculations). This note also demonstrates that human uncertainty is not related to actual knowledge. Hence, the increase in uncertainty that often accompanies advances in knowledge should be expected, and not treated as a scientific aberration. (author)

[en] Understanding of what makes a control room situation difficult to handle is important when studying operator performance, both with respect to prediction as well as improvement of the human performance. Previous exploratory work on complexity showed a potential for prediction and explanation of operator performance. This report investigates in further detail the theoretical background and the structure of operator rated task complexity. The report complements the previous work on complexity to make a basis for development of operator performance analysis tools. The first part of the report outlines an approach for studying the complexity of the control room crew's work. The approach draws upon man-machine research as well as problem solving research. The approach identifies five complexity-shaping components: 'task work characteristics', 'teamwork characteristics', 'individual skill', 'teamwork skill', and 'interface and support systems'. The crew's work complexity is related to concepts of human performance quality and human error. The second part of the report is a post-hoc exploratory analysis of four empirical HRP studies, where operators' conception of the complexity of control room work is assessed by questionnaires. The analysis deals with the structure of complexity questionnaire ratings, and the relationship between complexity ratings and human performance measures. The main findings from the analysis of structure was the identification of three task work factors which were named Masking, Information load and Temporal demand, and in addition the identification of one interface factor which was named Navigation. Post-hoc analysis suggests that operator's subjective complexity, which was assessed by questionnaires, is related to workload, task and system performance, and operator's self-rated performance. (Author). 28 refs., 47 tabs

No abstract available

No abstract available

[en] Coupled liquid-structure oscillations of a container of pool-type LMFBR (liquid metal fast breeder reactor) subjected to horizontal excitations are studied for small motions. First, a variational principle is introduced for a perfect fluid and a linear elastic solid. Ignoring pumps and IHX's (intermediate heat exchangers), the container is treated as an axisymmetric structure containing liquid sodium considered as a perfect fluid. Discretized equations of motion are derived from the variational principle, using FEM for axisymmetric shell and BEM for liquid sodium. Free vibration analysis is worked to obtain natural frequencies and modes. Then, a modal analysis method is applied to give time history of the liquid-structure system subject to horizontal excitations resonant to the first bulging mode. Characteristic features of resonance are observed in deformations of an inner oblique wall. (author)

[en] The simplest and most widely used plate theory is the classical small deflection theory founded by Lagrange. This theory does not include the effect of temperature and any middle surface forces arising from the deformation of the plate are ignored. The problem of plates subject to temperature distribution are, however, of great practical importance and to treat this class of problems, a generalization of the Lagrange plate equation is necessary. The problem treated here is that of a circular plate resting on elastic foundation and subject to an arbitrary temperature distribution across its thickness and to uniform normal surface loading. (orig.)

No abstract available